Available online at Energy Procedia 4 (2011) Energy Procedia 00 (2010) GHGT-10

Transcription

1 Available online at GGT-10 Synthesis and selection of hindered new amine absorbents for CO 2 capture Firoz Alam Chowdhury 1 *, iromichi Okabe, idetaka Yamada, Masami Onoda and Yuichi Fujioka Research Institute of Innovative Technology for the Earth (RITE), 9-2 Kizugawadai, Kizugawa-shi, Kyoto , Japan Elsevier use only: Received date here; revised date here; accepted date here Abstract This work focuses on synthesizing and selecting hindered new amine absorbents to reduce the regeneration energy cost for CO 2 capture. To achieve this target we synthesized seven secondary and two tertiary amine based CO 2 absorbents with systematic modification of their chemical structures by an appropriate placement of substituent functional groups especially the alkyl functions (e.g. methyl, isopropyl, isobutyl, secondary butyl etc.), relative to the position of the amino group. Performances evaluations of these new absorbents in aqueous solutions were carried out based on their CO 2 absorption rate, absorption capacity and heat of reaction measurements. Particular attention was paid to absorbents with a potential for high absorption rate and low heats of reaction. The results for the synthesized amino alcohols were then compared with conventional absorbents AMP and MDEA which were chosen as the base case for all comparisons. We found three high performance absorbents with advantages of high absorption rate and low heats of reaction compared with AMP and MDEA. Absorbents higher CO 2 absorption rate and lower heat energy consumption characteristics will reduce the regeneration energy cost of CO 2 during stripping. c Elsevier Published Ltd. byall Elsevier rights reserved Ltd. Keywords: amine absorbent, CO 2 capture, absorption-regeneration, reaction rate, heat of reaction 1. Introduction uman activities contribute to climate change particularly the burning of fossil fuels, causes emissions of CO 2, the main gas commonly referred to as greenhouse gas causing global warming. To prevent global warming by the greenhouse effect it is crucial to develop energy saving absorbents for capturing and separating CO 2 from its large point sources [1]. CO 2 capture by chemical absorption using an aqueous solution of amine based absorbents is a common industrial process and has, in many cases, been found to be the most viable solution compared with other processes [2]. Furthermore, in the currently used systems more than half of the capture cost is caused by the absorbent regeneration. In order to make it practical in the near future, it is essential to reduce the absorbent regeneration cost by developing new amine absorbents. To achieve this target, we focused on the development of * Corresponding author. Tel.: ; fax: address: firoz@rite.or.jp doi: /j.egypro

2 2 202 F.A. Author Chowdhury name / et al unique amine based CO 2 absorbents which have a low heat of reaction, a fast absorption rate and a high capacity for carbon dioxide. During chemical absorption process the acid gas is absorbed into the solution at lower temperatures and regenerates from the solution by heating to higher temperatures. The process chemistry is complex, but main reactions taking place are CO 2 Absorption: 2 R R' + CO 2 RR' RR'-COO - (R = or alkyl, R' = alkyl) Regeneration: RR'-COO - + RR' (more eat) CO RR' An unhindered amine forms a weakly bonded intermediate called carbamate that is fairly stable. Only half a mole of CO 2 is absorbed per mole of amine, as shown in the CO 2 absorption equation above. On application of heat, this carbamate dissociates to give back CO 2 and amine absorbent, as shown in the second equation above. Since the carbamate formed during absorption is quite stable, it takes lot of heat energy to break the bonds and to regenerate the absorbent. For other hindered amines (e.g., where R is a bulky group), the carbamate formed is not stable, and an alternate reaction leads to formation of bicarbonate ions and hence a higher theoretical capacity of one mole of CO 2 per mole of amine, as shown in the CO 2 absorption equation below CO 2 Absorption: RR'R" + CO O RR'R" + + CO 3 - (R = or alkyl, R', R" = alkyl) Regeneration: RR'R" + + CO (less eat) CO O + RR'R" The regeneration of these amines requires lesser amount of heat energy as compared with unhindered amines. Low rates of CO 2 absorption, however, make hindered amines difficult to use for removing CO 2 gas. In order to resolve the above mentioned limitations formulated solvents consists of amine blends and some chemical additives are receiving considerable attention. The advantages derived from amine blends are also limited to commercially available individual amines. The aforementioned facts motivate us to synthesis hindered secondary and tertiary alkanolamine based CO 2 absorbents that have a low heat of reaction and a fast absorption rate. 2. Research Target In the previous studies, commercially available amine absorbents were broadly investigated by a number of researchers. For example, the relation between absorption/desorption rates and amine structure was investigated from the screening tests (Bonenfaout et al., 2003) [3]. Measurements of vapor-liquid equilibrium (VLE) of amine solvent and heat of reaction were also studied (Jou et al., 1994, Mathonat et al., 1998, Mimura et al., 2005) [4-6]. Recently, a new class of amines referred to as sterically hindered amines, such as 2-amino-2-methyl-1-propanol (AMP), and 2-Piperidineethanol (2-PPE) have been proposed as commercially attractive new solvents for acid gas treating over commercial amines such as monoethanolamine (MEA), diethanolamine (DEA), 2-(ethylamino)ethanol, (EAE), and -methyldiethanolamine (MDEA) [7-8]. There is very little information available regarding synthesized new amines or amino alcohols. Recently, new types of amino alcohols were synthesized by Maneeintr et al [9] was applied to promote CO 2 capture performance. 6 5 MEA Absorption Rate (g-co2/l/min.) Preferred Target DEA MDEA AMP Reaction eat (kj/mol-co2) Figure 1. The trade-off between reaction heat and CO 2 absorption rate

3 F.A. Chowdhury Author name et al Development of new absorbents is an ongoing specific research task of the RITE projects. Past research of our group has developed several cost saving new absorbents for CO 2 capture [10-11]. They were evaluated on their CO 2 loading capacity, heats of reaction and absorption rates by comparison to MEA and MDEA. Those results showed that there is a structure-performance relationship between amino alcohols and CO 2 capture performance. The relationship between CO 2 absorption rate and heat of reaction, as determined experimentally, for a primary MEA and AMP, a secondary DEA and a tertiary MDEA alkanolamines are shown in Figure 1. It is apparent from Figure 1 that the heat of reaction and absorption rate of alkanolamines is dependent on the nitrogen substituent feature as MEA reacts faster than AMP, then DEA and MDEA. Figure 1 also shows hindered amines such as AMP and MDEA requires less heat energy and low rates of CO 2 absorption compared with unhindered amine MEA. Our research target was thus to increase the absorption rate and decrease the heat of reaction for hindered secondary and tertiary amines compared with conventional amines AMP and MDEA. 3. Experimental, Absorbent Selection Methodology, Results and Discussion 3.1. Experimental Chemicals All conventional amines (MEA, AMP, DEA and MDEA) were purchased from Sigma Aldrich chemical Co. and Tokyo Kasei Kogyo Co. Ltd. and were used without any further purification. The hindered new amino alcohols were synthesized in our laboratory by the alkylation reaction of amines and their corresponding alkylhalides. Synthesized amines purity and structure determination were established by GC and MR spectroscopy. For all amines a 30wt% aqueous solution was used for the screening test, vapor-liquid equilibrium test and heat of reaction measurements. 3-2 Absorbents selection methodology, Results and Discussion Synthesis and selection of new absorbents is very laborious. A tremendous amount of experimental work has to be done on characterizing the new solvents with respect to different properties. The hindered amino alcohols have been synthesized based on an approach of rational molecular design and synthesis. This involved a systematic tuning of their chemical structures by an appropriate placement of substituent functional groups, especially the alkyl function (e.g. methyl, isopropyl, isobutyl, secondary butyl etc.), relative to the position of the amino group. Some of the resulting amino alcohols were 2--methylamino-2-methyl-1-propanol (MAMP), 2--ethylamino-2-methyl-1- propanol (EAMP), 2-(isopropylamino)ethanol (IPAE), 2-(isobutylamino)ethanol (IBAE), 2- (secondarybutyamino)ethanol (SBAE), 2-(isopropyl)diethanolamine (IPDEA) and 1-Methyl-2-piperidineethanol (1M-2PPE) are listed in Table 1 and their chemical structures are shown in Figure 2. These new amino alcohols were designed to promote CO 2 capture performance and to study the effect of such placement of functional groups on the performance of the amino alcohols for CO 2 capture. Table 1 Investigated amine absorbents Conventional amine absorbents 1. 2-aminoethanol (MEA) 2. 2-amino-2-methyl-1-propanol (AMP) 3. Diethaholamine (DEA) 4. Methyldiethanolamine (MDEA) Synthesized amine absorbents methylamino-2-methyl-1-propanol (MAMP) ethylamino-2-methyl-1-propanol (EAMP) 7. 2-(isopropylamino)ethanol (IPAE) 8. 2-(isobutylamino)ethanol (IBAE) 9. 2-(secondarybutyamino)ethanol (SBAE) (isopropyl)diethanolamine (IPDEA) Methyl-2-piperidineethanol (1M-2PPE) To select unique absorbents from the above mentioned amines three fundamental experiments were performed in our laboratory to evaluate their CO 2 capture performance. They were screening tests, vapor-liquid equilibrium tests and heat of reaction measurements. Through these experiments solvent characteristics such as absorption-

4 4 204 F.A. Author Chowdhury name / et al regeneration rate, capacity of CO 2 capture, and heat of reaction were obtained. A parallel solubility study of conventional alkanolamines were also conducted at the same conditions for comparison of their performance in terms of absorption rate, absorption capacity and reaction heat. The details of these tests will be described in the following paragraphs. Figure 2. Investigated amine absorbents and their chemical structures Screening The goal of screening test was to clarify the initial reactivity of absorbents with CO 2. Figure 3 demonstrate the schematic diagram of the experimental apparatus. The equipment designed to operate at atmospheric pressure and temperatures up to 100 C consists of six bubble absorbers. Firstly, a 250ml glass scrubbing bottle filled with 50 ml of a 30wt% screening solvent was placed in the water bath controlled at 40 C. A 20% CO 2 gas balanced with 2 was then supplied to the bottle at a flow rate of 700ml/min. After 60 minutes of CO 2 absorption, the bottle was moved to the other water bath which was controlled at 70 C and CO 2 was regenerated from the solvent for 60 minutes. The flow rate and CO 2 concentration of the feed gas were constant for the absorption and regeneration tests. During the test, the outlet gas from the reactor was analyzed with a carbon dioxide analyzer (VA-3001, ORIBA). The CO 2 loading and absorption/regeneration rates with time were estimated from the measured CO 2 concentration. The reproducibility of the experiments was checked, and the error in all of the experimental measurements was found to be less than 3%. CO 2 absorption and regeneration of the four conventional amine absorbents (1-4) and the seven newly synthesized amine absorbents (5-11) were measured. Figure 4 expresses the typical example of screening tests. The saturation is reached for all compounds within the experimental time frame (one hour), enables a comparison of rates of absorption and saturation values. The absorption rates of all tested amines were calculated from 50% saturated CO 2 loading (absorption curves approach linearity in this range). The screening test provided a saturated CO 2 loading capacities and absorption/desorption rates of amines during a chemical absorption process. The relative performances of all amines are shown in Table 2 where AMP and MDEA are regarded as a reference absorbent for performance comparison. Gas supply CO 2 20% 2 80% Flow rate 700 ml/min Conventional amine absorbents (1-4) 2 Water bath for absorption (40 C) 2 CO 2 analyzer After 60min O Water bath for regeneration (70 C) 1. MEA 2. AMP 3. DEA Synthesized amine absorbents (5-11) Figure 3. Schematic diagram of screening test apparatus O 5. MAMP 6. EAMP 7. IPAE 8. IBAE 9. SBAE 10. IPDEA 11. 1M-2PPE Absorption amount (g-co2/l-soln) 4. MDEA wt%_AMP 30wt%_SBAE 30wt%_IBAE 30wt%_MDEA Absorption at 40, 30wt%_IPAE 30wt%_1M-2PPE 30wt%_IPDEA Regeneration at 70, 1h Time (min) Figure 4. Typical examples of screening tests

5 F.A. Chowdhury Author name et al Figure 4 demonstrate the saturated CO 2 loading and absorption rate of IPAE and 1M-2PPE were much higher than AMP and MDEA respectively. In case of developing new absorbents we thus focused on the absorbents that outperformed than AMP and MDEA in screening tests. Table 2 shows the experimental results of all screening amines with regards to their absorption rates, saturated CO 2 loading and absorption capacities. Table 2 and Figure 4 clearly show that the absorption rates of synthetic secondary amine IPAE and IBAE are higher than the absorption rate of AMP (unhindered MEA doesn t considered for comparison). Similar effect was also observed in case of synthetic tertiary amine IPDEA and 1M-2PPE in comparison to MDEA shown in Table 2 and Figure 4. In addition synthetic amines also present reasonable CO 2 absorption capacity entry 7, 9, 10 and 11 (Table 2) in comparison to AMP and MDEA. The slow absorption rate of the -substituted, -dimethylated amines MAMP and EAMP relative to AMP indicates that the presence of three bulky alkyl groups around the reaction site causes too much of a physical barrier thus greatly decreasing the reaction rate. Screening results provided clear trends concerning the structural effects of hindered amine absorbents. Methyl, Isopropyl and Isobutyl groups were found to be the most suitable substituted functional groups for the enhancement of initial absorption rate and capacity. From the screening test results several amine absorbents with high CO 2 absorption rates and capacities compared to AMP and MDEA were selected for vapor-liquid equilibrium test. Table 2. Experimental results for screening and heat of reaction tests Amine Absorbents Absorbents 30wt% aqueous solutions Absorption rate a g-co 2/L-soln./min Absorption amount b g-co 2/L-soln. Regeneration amount c g-co 2/L-soln. Absorption capacity d g-co 2/L-soln. Reaction eat e kj/mol-co 2 1 MEA AMP DEA MDEA MAMP EAMP IPAE IBAE SBAE IPDEA M-2PPE a Absorption rate calculated at 50% of total CO2 loading, b Maximum CO 2 loading at 40, 1h, c Maximum regeneration at 70, 1h, Difference of CO2 loading between (40 ~70 ), e Reaction heat measured between = (0.4 ~ 0.6) mol-co 2/mol-amine at 40 Vapor-liquid equilibrium (VLE) Figure 5 shows the experimental apparatus used for vapor-liquid equilibrium tests which contained a 700-cm 3 crstal glass cylindrical vessel (autoclave), a water-saturator, an electric heater, a mechanical stirrer, a condenser and a CO 2 analyzer. It was designed to operate at temperatures up to 200 C and pressures up to 1 MPa. Tests were conducted between 40 C and 120 C. Before starting the experiment the apparatus was purged with 2. During the experiment the autoclave was filled with a test solvent and heated with an electric heater. The gas which was controlled to a specific CO 2 concentration was then supplied to the autoclave after flowing through a water-saturator. The equilibrium was determined when the CO 2 analyzer indicated a constant CO 2 concentration in the outlet gas. This typically took 2 to 3 hours. To analyze the equilibrium condition the CO 2 concentration in both gas and liquid phase was measured. The CO 2 partial pressure was derived from the temperature, total pressure and the measured CO 2 concentration. For the liquid phase a sample was drawn from the autoclave and the amount of absorbed CO 2 was measured with a Total Organic Carbon analyzer (TOC-VC, Shimadzu). Solvents that showed superior performance from screening tests were selected for the evaluation of their vaporliquid equilibrium property. Figure 6 express the typical example of vapor-liquid equilibrium tests for conventional amino alcohols MEA, MDEA and synthetic amino alcohols IPAE and 1M-2PPE. The effective CO 2 loading of all the chosen absorbents are shown in Table 3. The effective CO 2 loading is the difference between absorber and

6 6 206 F.A. Author Chowdhury name / et al regeneration conditions of the CO 2 capture system. The effective CO 2 loading, is a key index that determines the net cyclic capacity in the CO 2 capture system. Back pressure regulator CO2 analyzer Pressure gauge Sampling line Condenser Temperature sensor M Mass flow controller Mass flow Saturator controller Crystal glass eater 2 cylinder CO2 cylinder Figure 5. Experimental apparatus used for vapor-liquid equilibrium tests Figure 6. Typical examples of vapor-liquid equilibrium (VLE) tests The experimental results of the selected four new absorbents compared to conventional amines at various conditions are presented in Table 3. Data from Table 3 shows that conventional absorbents AMP and hindered new absorbents IPAE and 1M-2PPE have good VLE property. This means, these three absorbents absorb more CO 2 and release more CO 2 off at high temperature. Good VLE property will reduce liquid flow rate and steam required for CO 2 recovery. This advantage can result in lowering the cost for gas treating process thereby making it more economically feasible. Table 3. Experimental results for VLE tests Amine Absorbents CO 2 Loading (g-co 2 /L-soln.) Effective Absorbents 30wt% aqueous solutions Absorber condition a Regeneration condition b CO 2 loading c g-co 2 /Lsoln. 1 MEA AMP DEA MDEA IPAE IBAE IPDEA M-2PPE a Temperature 40, CO 2 partial pressure 20kPa, b Temperature 120, CO 2 partial pressure 100kPa, c Difference of CO 2 loading between 40 (20kPa) and 120 (100kPa) under equilibrium condition eat of Reaction The heat of reaction for selected absorbents was measured with a Differential Reaction Calorimeter (DRC, SETARAM) as shown in Figure 7. This instrument works on the simple principle of differential thermal analysis which continuously measures the difference in temperature T between a sample reactor and a reference reactor. The amount of absorbed CO 2 in the solvent was then measured with a Total Organic Carbon analyzer. To calculate

7 F.A. Chowdhury Author name et al the heat of reaction the total generated heat was divided by the increase in CO 2 in the solvent during CO 2 injection. The solvent volume was 150ml and the injected gas was 100% CO 2. Figure 7. Experimental apparatus for heat of reaction measurements (SETARAM, DRC) 100 Reaction eat (kj/mol-co2) MEA AMP DEA MDEA IPAE IBAE SBAE IPDEA 1M-2PPE Figure 8. Experimental results for the heat of reaction tests We selected nine test absorbents (four conventional and five synthetic amines) for the heat of reaction measurements. The results are presented in Figure 8 and Table 2. The general order of heat of reaction for conventional amines became primary>secondary>tertiary as expected from previous measurements (Figure 1). Table 2 also shows that the heat of reaction for synthesized amino alcohols provide much lower reaction heat than that of the conventional amine. Synthetic secondary amine compared with AMP, the heats of reaction as follows, AMP>SBAE>IPAE, IBAE and that for tertiary amines were: MDEA, IPDEA>1M-2PPE. Figure 9 shows the relationship between heat of reaction and CO 2 absorption rate between conventional amines and synthesized amino alcohols. As mentioned in Figure 1, there was a trade-off relationship between heat of reaction and absorption rate for primary, secondary and tertiary amines. That is, the absorbents of low heat of reaction usually had the feature of low CO 2 absorption rate. owever, the results in this study showed that IPAE, IBAE, IPDEA and 1M-2PPE solvents were plotted at outside of the trade-off relationship formed by the conventional amine absorbents. amely, IPAE, IBAE and 1M-2PPE solvents had a unique performance that they had a feature of a low heat of reaction and kept a moderately high absorption rate.

8 8 208 F.A. Author Chowdhury name / et al Absorption Rate (g-co2/l/min.) IPAE IBAE Preferred Target DEA 1M-2PPE IPDEA SBAE MDEA AMP MEA Reaction eat (kj/mol-co2) Figure 9. Performance comparisons between conventional amines and synthesized new amines 4. Conclusions We demonstrated that by way of rational molecular design and placement of functional groups, hindered amino alcohols for promoting CO 2 capture can be developed. It has been shown that the placement of functional groups within the amino alcohols affects the performance of the amino alcohols in CO 2 capture. Thus, there is a structureperformance relationship between amino alcohols and CO 2 capture performance. We have investigated seven hindered secondary and two tertiary alkanol amine based CO 2 absorbents with different chemical structures and found two secondary (IPAE and IBAE) and two tertiary (1M-2PPE and IPDEA) alkanol amines with higher CO 2 absorption rates and lower heat of absorption than the conventional absorbents AMP and MDEA. In addition new absorbents have reasonable performance in terms of cyclic capacities. Absorbents higher CO 2 absorption rate and lower heat energy consumption characteristics will reduce the regeneration energy cost of CO 2 during stripping. Finally, this can lead to a lower cost for CO 2 capture. Acknowledgements This study was conducted as a part of the COCS project, which was financially supported by the Ministry of Economy, Trade and Industry (METI), Japan and carried in collaboration with four Japanese companies, ippon Steel Co., ippon Steel Engineering Co., Ltd., Mitsubishi eavy Industries, Ltd. and The Kansai Electric Power Co., Inc. References [1] Rao, A. B.; Rubin, E. S Environ. Sci. Technol. 2002, 36: [2] Meisen, A.; Shuai, X. Convers. Manage. 1997, 38, S37. [3] Bonenfant, D., M. Mimeault, M; ausler; R. Ind. Eng. Chem. Res., 2003, 42, [4] Jou, F. Y.; Otto, F. D.; Mather, A. E. Ind. Eng. Chem. Res., 1994, 33, [5] Mathonat, C.; Majer, V.; Mather, A. E.; Grolier, J. P. E. Ind. Eng. Chem. Res., 1998, 37, [6] Mimura, T.; Yagi, Y.; Takashina, T.; Yoshiyama, R.; onda, A. Kagaku Kogaku Ronbunshu, 2005, 41(4), [7] Yoon, J..; Baek, J. I.; Yamamoto, Y.; Komai, T.; Kawamura, T. Chem. Eng. Sci. 2003, 58, [8] Paul, S.; Ghoshal, A. K.; Mandal, B. Ind. Eng. Chem. Res., 2009, 48, [9] Maneeintr, K.; Idem, O. R; Tontiwachwuthikul, P.; Wee. G. A. 1, 2009, [10]. Chowdhury F. A., Okabe., Shimizu S., Onoda M., Fujioka Y. 1, 2009, [11]. Goto K., Okabe., Shimizu S., Onoda M., Fujioka Y. 1, 2009,